Phenformin

PHENFORMIN AND LACTIC ACIDOSIS: A CASE REPORT AND REVIEW
Shun C. Kwong, MD* and Jeffrey Brubacher, MD, FRCPC (EM)†
*Department of Emergency Medicine, St. Luke’s–Roosevelt Hospital, New York, New York, and †Department of Emergency Medicine, Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia, Canada
Reprint Address: Shun Kwong, MD, Department of Emergency Medicine, Methodist Hospital, 8303 Dodge Street, Omaha, NE 68114

e Abstract—Phenformin was removed from the U.S. mar- ket 20 years ago because of a high incidence of lactic acidosis. Unfortunately, this medication is still available from foreign sources. Another biguanide, metformin, was reintroduced to the United States market for the treatment of diabetes. Biguanide-induced lactic acidosis should be included in the differential diagnosis of elevated anion gap metabolic acidosis. We present a case of phenformin-in- duced lactic acidosis in which we were consulted at the local poison control center. We also review its pathophysiology, presentation, and treatment. A review of the actions of phenformin illustrates the mechanism of pathology that may also occur with metformin. Risk factors for the devel- opment of lactic acidosis include renal deficiency, hepatic disease, cardiac disease, and drug interaction such as cime- tidine. © 1998 Elsevier Science Inc.

e Keywords— biguanides; lactic acidosis; phenformin

INTRODUCTION

In medieval times, the French lilac (Galega officinalis), also known as goat’s rue, was used in southern and eastern Europe to treat diabetes. Guanidine, the active ingredient in French lilac, was found to possess hypo- glycemic activity in 1918. Guanidine’s clinical use was limited by its toxicity, which includes hypotension, atrial fibrillation, tremor, ataxia, and seizures. Biguanides were developed from guanidine and investigated for the treat-

ment of diabetes in the 1920s. However, interest in these compounds waned with the discovery of insulin in 1921 (1,2).
The biguanide phenformin was introduced in the U.S. in 1957 to treat non-insulin-dependent diabetes mellitus (NIDDM). Phenformin was withdrawn from the U.S. and European markets in 1977 because of the high incidence of associated lactic acidosis occurring at therapeutic doses. The therapeutic dose of phenformin was 50 to 200 mg per day, and patients developing lactic acidosis took an average of 123 mg per day (7). Phenformin remains available in other parts of the world. Another biguanide, metformin, has been used extensively in Canada and Europe for over a decade and was introduced to the U.S. in 1995.
Biguanides are currently used to overcome insulin resistance, especially in obese NIDDM patients who fail diet and exercise therapy. Insulin resistance in peripheral tissues results in impaired uptake and utilization of glu- cose, and decreased glycogen formation (3). In the liver, insulin resistance results in increased gluconeogenesis and elevated basal hepatic glucose output (4). Biguanides are similar in antihyperglycemic efficacy to sulfonyl- ureas and can be used either alone or as an adjunct to other therapy (1,5).
All biguanides can cause lactic acidosis, but the inci- dence is highest in patients using phenformin. One in 4,000 patients taking phenformin develops lactic acido-

Pharmacology in Emergency Medicine is coordinated by Richard Clark, MD, of the University of California, San Diego
Medical Center, and the San Diego Regional Poison Center, San Diego, California
RECEIVED: 25 April 1997; FINAL SUBMISSION RECEIVED: 5 January 1998;
ACCEPTED: 16 January 1998
881

sis, compared with one in 40,000 to 80,000 in patients taking metformin or buformin (6). In a review of 330 cases of biguanide associated lactic acidosis, 285 cases involved phenformin, and 50.3% of the patients died (7). Biguanide toxicity should be considered in the differen- tial diagnosis of diabetics who present with lactic acido- sis.

CASE REPORT

A 67-year-old diabetic male who had been started on oral hypoglycemics in China two years previously presented with a chief complaint of weakness and possible loss of consciousness 30 min before arrival at the emergency department (ED). The patient had been in his usual state of health until the evening before presentation when he had an episode of diarrhea for which he took an extra dose of his diabetic medication. The next morning, he developed vomiting and anorexia but was able to con- tinue taking his medications. By that evening, the patient had become lethargic and confused, and a neighbor called the emergency medical services (EMS). The EMS technicians found the patient to be diaphoretic and le- thargic after exercise, with the following vital signs: blood pressure 200/100 mmHg, pulse 96 beats/min, re- spiratory rate 18 breaths/min. The patient was oriented to name only. No seizure activity was reported. No re- sponse was noted by the EMS technicians after the administration of one ampule (50 mL) of D50W. In the ED, the patient was awake but confused, with a finger stick glucose of 20 mg/dL. The patient became fully oriented after being given an additional two ampules of D50W. He denied chest pain, nausea, palpitations, dizzi- ness, shortness of breath, or abdominal pain.
The patient had been previously healthy except for a 2-year history of NIDDM treated with medications ob- tained from China. He was taking phenformin and glib- enclamide, a sulfonylurea. He had no allergies, used alcohol only occasionally, and did not use tobacco. On initial physical examination in the ED, he was in no apparent distress, confused, and easily arousable. The rest of the physical examination was unremarkable.
On presentation, the complete blood count was: white blood cells 15,800/mm3, hemoglobin 13.4 G/dL, hemat- ocrit 41%, and platelet count 297,000/mm3. The sodium was 144 mEq/L, potassium 3.9 mEq/L, chloride 112 mEq/L, bicarbonate 10.4 mEq/L, glucose 10 mg/dL, blood urea nitrogen 26 mg/dL, creatinine 1.6 mg/dL, anion gap 21.6 mEq/L, serum osmolality 318 mosm/L, and osmolarity gap 20 mosm/L. Ethanol, aspirin, ethyl- ene glycol, methanol, and paraldehyde were absent from toxicology screening. A reference laboratory determined that phenformin was qualitatively present by HPLC with

UV detection (National Medical Services, 800-522- 6671). The blood was not analyzed for ketones. Urinal- ysis showed glucose 250 mg/dL, ketones 15 mg/dL, trace blood pH 5, and total protein 15 mg/dL. Urine electro- lytes showed sodium 130 mEq/L, potassium 15 mEq/L, chloride 23 mEq/L, creatinine 17.5 mg/dL, and osmola- lity 400 mosm/L. The first lactic acid level was 24.6 mmol/L (normal <1.6 mmol/L), which increased to 30 mmol/L. The initial room air arterial blood gas was pH 6.91, pCO2 32 mmHg, pO2 110 mmHg. The electrocar- diogram showed sinus tachycardia of 106, normal axis with normal QRS duration, and no ischemic changes. The chest radiograph and computed tomography (CT) scan of the brain were normal. The patient’s presenting hypoglycemia was thought to be the result of vomiting, anorexia, poor oral intake, and the presence of gliben- clamide. Although the patient initially appeared to be in no acute distress and was easily arousable, he apparently deteriorated. He was intubated for airway protection and to maintain hyperventilation. Activated charcoal was given. Sodium bicarbonate was started with a total of 19 ampules given over 12 h. Hemodialysis was recom- mended by the poison control center but not performed since the patient was improving in the intensive care unit. By 12 h after presentation, the lactate had fallen to 5.6 mEq/L and the blood gases were: pH 7.27, pCO2 23 mmHg, pO2 104 mmHg. At time of discharge 6 days later, the acid-base status had normalized, and the patient recovered. DISCUSSION Increased basal hepatic production is the primary factor responsible for increased fasting glucose concentrations in obese diabetics, and the main beneficial action of biguanides is inhibition of hepatic gluconeogenesis (9,10). Another beneficial effect of biguanides may be increased intestinal usage of glucose. This effect is seen in obese rats, normal mice, and STZ-induced diabetic mice (1). Increased glucose uptake by the red blood cells (RBC) may be another important biguanide action. In- vitro studies of RBC from IDDM patients under hyper- glycemic conditions showed increased formation of glu- cose 6 phosphate and increased glucose utilization after biguanide administration (1). Biguanides help in the treatment of obese patients with insulin-resistant NIDDM and add another therapeu- tic option when these patients fail diet and exercise therapy. Non-obese NIDDM patients may benefit as well (8). Biguanides result in decreased insulin resistance with insignificant effects on insulin levels and little or no effect on glucagon, somatostatin, growth hormone, or cortisol levels (1). In contrast to sulfonylureas, bigua- nides do not usually cause hypoglycemia, even in over- dose (11,12), and changes in glucose levels are hardly apparent in nondiabetic volunteers unless glucose con- centrations are artificially raised (13). Biguanide toxicity induces anorexia, nausea, and vomiting, resulting in a fasting state with a subsequent fall in insulin levels (14). Basal levels of insulin are sufficient to prevent catabolism of fat and muscle in normal persons but not in fasting diabetics (15). Protein catabolism results in increased levels of alanine and other glucogenic amino acids that can be converted to pyruvate to be used as substrates for gluconeogenesis (16). With gluconeogenesis inhibited by biguanides, pyruvate accumulates. Oxidation of the fatty acids re- leased from fat catabolism depletes NAD+ and increases NADH (17). Inhibition of oxidative phosphorylation by phenformin impairs the ability of the mitochondria to generate NAD+ from NADH, further increasing the NADH/NAD+ ratio (15). The resultant increased NADH/NAD+ ratio inhibits pyruvate dehydrogenase and the entry of pyruvate into the Kreb’s cycle (15). Fatty acid oxidation increases the acetyl CoA/CoA ratio, which further decreases the entry of pyruvate into the Kreb’s cycle (18). With pyruvate dehydrogenase inhib- ited and gluconeogenesis blocked, the accumulated pyru- vate is metabolized to lactate; this reaction is favored by the increased NADH/NAD+ ratio (19). Fatty acid me- tabolism also yields ketone bodies (acetoacetate and be- ta-hydroxybutyrate) that further exacerbate the acidosis (Figure 1). Various risk factors for toxicity should be considered. Renal deficiency can cause increased levels of both phenformin and metformin at therapeutic doses (15,20). Phenformin decreases the glomerular filtration rate and impairs the ability of the kidneys to excrete an acid load (21). Cimetidine reduces renal clearance of metformin (22). Patients with hepatic dysfunction will have im- paired phenformin and lactate metabolism (20). Bigua- nides are negative inotropes, decreasing cardiac output and hepatic blood flow and further decreasing the hepatic clearance of lactate (23,24). Other factors contributing to lactic acidosis include a previous history of lactic acido- sis, cardiac insufficiency, hypoxia, and alcohol abuse because of possible liver impairment and to alcohol’s hypoglycemic effect (25). Clinical Toxicity The hallmark of biguanide-associated toxicity is severe lactic acidosis not explained by hypoperfusion or hyp- oxia (20). Biguanides increase the plasma lactate con- centration, increase the entry of lactate into plasma, Figure 1. Biguanide acidosis. Insulin of deficiency and accu- mulation of pyruvate are central to the development of lactic acidosis. Insulin deficiency results in protein catabolism (1) and fat catabolism (2). Protein catabolism results in in- creased pyruvate (3). Biguanides inhibit pyruvate carboxy- lase (4) and therefore prevent the use of pyruvate in glucone- ogenesis. Pyruvate entry into the Kreb’s cycle (5) is inhibited because of excess acetyl CoA from fat catabolism (6) and from inhibition of the Kreb’s cycle itself (7), which occurs secondary to excess NADH from fat catabolism (8) and in- hibition of gluconeogenesis. This NADH excess also favors the formation of Beta-hydroxybutyrate from acetoacetate. With gluconeogenesis blocked by biguanides and entry of pyruvate into the Kreb’s cycle inhibited, the accumulated pyruvate is metabolized to lactate (9). Ketone formation from fat catabolism adds to the acidosis (10). inhibit lactate oxidation, impair oxidative phosphoryla- tion, and increase the release of lactate from muscle (15,26 –28). In contrast to anaerobic lactic acidosis where cellular hypoxia is involved, biguanide-associated lactic acidosis involves increased lactate production and decreased lactate metabolism in an otherwise-aerobic state. The signs and symptoms of biguanide toxicity are nonspecific and include vomiting, somnolence, nausea, epigastric pain, anorexia, hyperpnea, lethargy, diarrhea, and thirst (20). The mortality rate may be as high as 50% overall and is correlated with increasing age, shock, and severity of metabolic acidosis (7,33). Together with the history and impressively elevated lactate levels, meta- bolic acidosis from biguanides may be distinguished from the acidosis of diabetic ketoacidosis (DKA) where ketosis predominates. In a review of 76 published cases of phenformin- associated lactic acidosis by Dembo et al., the typical presentation consists of a brief prodrome of prominent gastrointestinal symptoms with severe metabolic acido- sis (15). Nausea with or without vomiting is the most common symptom, occurring in 83% of cases, followed by Kussmaul respiration and altered levels of conscious- ness in 78% and 70%, respectively. Other clinical find- ings include dehydration, hypothermia, and hypotension. In Luft’s review of 330 patients with biguanide-induced lactic acidosis, the most common complaints were nau- sea, vomiting, somnolence, and epigastric pain (7). A decreased level of consciousness was present in 177 of the 212 patients for whom mental status was docu- mented. The patients with the lowest blood glucose and the highest blood urea nitrogen had the most severely impaired levels of consciousness. Cardiovascular shock was described in 13 patients as the first sign of lactic acidosis. Although not explicitly stated in the article, shock presumably means hypotension with tissue hypo- perfusion. Another illness, such as cardiovascular dis- ease (44%), renal insufficiency (35%), or an infectious process (25%), was present in 214 patients. Seventeen patients had taken large amounts of alcohol before lactic acidosis developed (7). In Dembo’s review, 97% of patients had weakly pos- itive or negative plasma ketones, 90% had impaired renal function (BUN > 20 mg/dL, creatinine > 1.4 mg/dL), and 66% had a pH of 7.1 or less (15). Serum ketones were believed to be artificially low since the sodium nitroprusside test for ketones only detects acetoacetate and acetone, not B-hydroxybutyrate. This belief was supported by the fact that 80% of the patients had un- measured anions > 15 mEq in excess of blood lactate, a clue to increased beta-hydroxybutyrate. Renal insuffi- ciency may also be a factor in elevation of unmeasured anions. In Luft’s series, the typical acid-base status was a decompensated metabolic acidosis with a marked ven- tilatory response (7). Eighty-one percent had pO2 above 80 mmHg, which demonstrates adequate oxygenation. The mean lactate was 16.9 mmol/L overall, 18.5 mmol/L in fatal cases, and 15.6 mmol/L in survivors. The mean pH was 6.95 overall, 6.91 in fatal cases, and 7.00 in survivors.
Several factors explain the higher incidence of lactic acidosis with phenformin than with metformin. Met- formin requires higher blood levels than phenformin to cause lactic acidosis and has a wider therapeutic window (7,20,29). Phenformin undergoes hepatic metabolism and renal elimination with a half-life of 12 h whereas metformin is excreted unchanged by the kidneys with a half-life of 1.5 h (20). Metformin, unlike phenformin, does not bind well to mitochondrial membranes and does not inhibit the electron transport chain (1). Furthermore, some people have an inherited inability to hydrolyze phenformin, resulting in increased phenformin levels and a greater risk of lactic acidosis (30 –32).

Treatment

Initial treatment of biguanide toxicity is directed at the support of ventilation and circulation. Gastrointestinal decontamination should be considered if a recent inges- tion is suspected. Blood should be analyzed to determine

Figure 2. Effect of insulin on lactic acidosis.

blood gases and lactic acid level. Other causes of meta- bolic acidosis, especially DKA, should be ruled out. Intubation and hyperventilation will provide respiratory support and quickly improve the pH.
Biguanide-induced lactic acidosis can be treated with insulin and glucose. Animal studies show that this treat- ment is effective (34). In a retrospective analysis of nonrandomized human series, insulin appeared to in- crease survival (15,35). As outlined earlier, insulin defi- ciency worsens biguanide toxicity by increasing the for- mation of pyruvate from protein catabolism, by indirectly increasing the NADH/NAD+ ratio, and by inhibiting pyruvate dehydrogenase. These factors all fa- vor the accumulation of pyruvate and its subsequent conversion to lactate. Insulin deficiency also causes a superimposed ketoacidosis secondary to catabolism of fat. Insulin and glucose doses used to treat biguanide- induced lactic acidosis average 10 –20 units of insulin and 5–12.5 g of glucose every 4 h (Figure 2; Reference 35).
Bicarbonate use is controversial. Its use presumes that increased blood pH is beneficial. Myocardial contractil- ity and responsiveness to catecholamines may decrease significantly at an arterial pH below 7.0, supporting bicarbonate use (20). Theoretical concerns about the use of bicarbonate include paradoxical intracellular and ce- rebrospinal fluid acidification, increased affinity of he- moglobin for O2 with subsequent decreased O2 delivery to tissues, hypernatremia, hypokalemia, and volume overload. Furthermore, bicarbonate increases cellular membrane permeability to biguanides, which ultimately increases lactate production (20). The decision to use bicarbonate must be made with these facts in mind. Many clinicians will give bicarbonate in the well-venti- lated patient with pH < 7.0. Hemodialysis will treat biguanide-induced lactic aci- dosis by removing lactate, acetoacetate, and beta hy- droxybutyrate. In addition, hemodialysis will remove metformin. Unfortunately, phenformin is not readily di- alyzable because of its large volume of distribution (20). In one study, only 1.5% of total phenformin ingested was removed by hemodialysis (36). Sodium dichloroacetate (DCA) is an experimental drug that increases the activity of pyruvate dehydroge- nase. This results in decreased intracellular lactate pro- duction and increased lactate metabolism. It can also directly increase cardiac output in experimental phen- formin-induced lactic acidosis (24,37). Another study found DCA to increase cardiac output and decrease cor- onary arterial lactate concentrations in patients with cor- onary artery disease (38). It prevents the development of, and partially reverses, lactic acidosis in diabetic rats given phenformin (39). However, preliminary studies of sodium dichloroacetate in humans show insignificantly increased survival from lactic acidosis caused by sepsis, shock, and cardiac arrest (40). A recent placebo-con- trolled clinical trial also shows clinically insignificant improvement in dichloroacetate-treated patients with lac- tic acidosis caused by sepsis (e.g., bowel infarction), cardiogenic shock, and hemorrhagic shock (38). In this study, “no restrictions were made regarding the cause (or causes) of lactic acidosis, except that patients with DKA were excluded” (38). Other details of the causes of acidosis were not made explicit. The use of this medi- cation in biguanide-induced lactic acidosis should be considered of unproven benefit at this time. CONCLUSION The risk of symptomatic lactic acidosis is small but real in patients treated with biguanides. The incidence of biguanide-induced lactic acidosis in the U.S. decreased markedly when phenformin was removed from the American market in 1977. Although less frequently than with phenformin, metformin can also cause lactic acido- sis, and this source can be expected to increase with the reintroduction of metformin into the U.S. market. Pa- tients may also obtain phenformin from foreign or un- controlled sources of medications entering the country, as occurred in the present case. The risk of developing lactic acidosis increases when other risk factors, such as coronary or renal insufficiency, or dehydration are present. Key factors in the history are the presence of NIDDM, biguanide use, and renal insufficiency. The laboratory analysis will show a marked anion gap metabolic acido- sis with an elevated lactate level. In addition to support- ive treatment, specific measures may include intubation and hyperventilation, insulin and glucose, and bicarbon- ate. Hemodialysis will correct the acidosis and may remove metformin but not phenformin. REFERENCES 1. Bailey CJ. Biguanides and NIDDM. Diabetes Care 1992;15:755– 72. 2. Bailey CJ, Day C. Traditional plant medicines as treatments for diabetes. Diabetes Care. 1989;553– 64. 3. Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman RG. Quantification of muscle glycogen synthesis in normal sub- jects and subjects with non-insulin-dependent diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med. 1990; 322:223– 8. 4. Consoli A, Nurjhan N, Capani F, Gerich J. Predominant role of gluconeogenesis in increased hepatic glucose production in NIDDM. Diabetes 1989;38:550 –7. 5. Bailey CJ, Turner RC. Metformin. N Engl J Med. 1996;334: 574 –9. 6. Kreisberg R, Wood BC. Drug and chemical-induced metabolic acidosis. Clin Endocr Metab. 1983;12:391– 411. 7. Luft D, Schmulling RM, Eggstein M. Lactic acidosis in biguanide- treated diabetics: a review of 330 cases. Diabetologia 1978;14:75– 87. 8. Clarke BF, Duncan LJ. Comparison of chlorpropamide and met- formin treatment on weight and blood glucose response of uncon- trolled obese diabetics. Lancet. 1968;1:123– 6. 9. DeFronzo RA, Bonadonna RC, Ferrannini E. Pathogenesis of NIDDM: a balanced overview. Diabetes Care 1992;15:318 – 68. 10. DeFronzo RA, Goodman AM, and the Multicenter Metformin Study Group. Efficacy of metformin in patients with NIDDM. N Engl J Med. 1995;333:541–9. 11. Sterne J, Junien JL. Metformin: pharmacological mechanisms of the antidiabetic and antilipidec effects and clinical consequences. R Soc Med Int Congr Symp Ser. 1981;48:3–16. 12. McLelland J. Recovery from metformin overdose. Diabetic Med. 1985;2:410 –1. 13. Hermann LS. Metformin: a review of its pharmacological proper- ties and therapeutic use. Diabetes Metab Rev. 1979;5:233– 45. 14. Cahill GF, Jr. Physiology of insulin in man. Diabetes 1971;20: 785–99. 15. Dembo AJ, Marliss EB, Halperin ML. Insulin therapy in phen- formin-associated lactic acidosis. Diabetes 1975;24:28 –35. 16. Mehler AH. Amino acid metabolism II. Metabolism of the indi- vidual amino acids. In: Devlin T, ed. Textbook of Biochemistry with Clinical Correlations. New York: John Wiley and Sons; 1986:454 –5. 17. McGarry JD. Lipid metabolism I. Utilization and storage of energy in lipid form. In: Devlin T, ed. Textbook of Biochemistry with Clinical Correlations. New York: John Wiley and Sons; 1986:377. 18. Olson MS. Bioenergetics and oxidative metabolism. In: Devlin T, ed. Textbook of Biochemistry with Clinical Correlations. New York: John Wiley and Sons; 1986:223–5. 19. Harris RA. Carbohydrate metabolism I: major metabolic pathways and their control. In: Devlin T, ed. Textbook of Biochemistry with Clinical Correlations. New York: John Wiley and Sons; 1986:273. 20. Gan SC, Barr J, Arieff AI, Pearl R. Biguanide-associated lactic acidosis. Arch Intern Med. 1992;152:2333– 6. 21. Rooth G, Bandman U. Renal response to acid load after phen- formin. Br Med J. 1973;4:256 –7. 22. Somogyi A, Stockley C, Keal J, Rolan P, Bochner F. Reduction of metformin renal tubular secretion by cimetidine in man. Br J Clin Pharmacol. 1987;23:544 –51. 23. Arieff AI, Park R, Leach WJ, Lazarowitz VC. Pathophysiology of experimental lactic acidosis in dogs. Am J Physiol. 1980;239: F135– 42. 24. Park R, Arieff AI. Treatment of lactic acidosis with dichloroacetate in dogs. J Clin Invest. 1982;70:853– 62. 25. Bailey CJ, Flatt PR, Marks V. Drugs inducing hypoglycemia. Pharmacol Ther. 1989;42:361– 84. 26. Kreisberg RA, Pennington LF, Boshell BR. Lactate turnover and gluconeogenesis in obesity: effect of phenformin. Diabetes 1970; 19:64 –9. 27. Searle GL, Siperstein MD. Lactic acidosis associated with phen- formin therapy: evidence that inhibited lactate oxidation is the causative factor. Diabetes 1975;24:741–5. 28. Williams RH, Steiner DF. Summarization of studies relative to the mechanism of the phenethylbiguanide hypoglycemia. Metabolism 1959;8:548 –52. 29. Assan R, Heuclin C, Girard JR, LeMaire F, Attali JR. Phenformin- induced lactic acidosis in diabetic patients. Diabetes 1975;24:791– 800. 30. Oates NS, Shah RR, Idle JR, Smith RL. Influence of oxidation polymorphism on phenformin kinetics and dynamics. Clin Phar- macol Ther. 1983;34:827–34. 31. Shah RR, Evans DA, Oates NS, Idle JR, Smith RL. The genetic control of phenformin 4-hydroxylation. J Med Genet. 1985;22: 361– 6. 32. Oates NS, Shah RR, Idle JR, Smith RL. Genetic polymorphism of phenformin 4-hydroxylation. Clin Pharmacol Ther. 1982;32: 81–9. 33. Fulop M, Hoberman HD. Phenformin-associated metabolic acido- sis. Diabetes 1976;25:292– 6. 34. Loubatieres A, Ribes G, Valette G. Reduction by sodium dichlo- roacetate and by insulin of the severe hyperlactemia following phenformin administration in anesthetized or awake dogs. CR Acad SciHebd Seances Acad Sci D. 1977:284: 325–7.
35. Misbin I. Phenformin-associated lactic acidosis: pathogenesis and treatment. Ann Intern Med. 1977;87:591–5.
36. Tobin M, Mookerjee BK. Hemodialysis for phenformin-associated lactic acidosis. J Dial. 1978;2:273– 85.
37. Stacpoole PW. The pharmacology of dichloroacetate. Metabolism 1989;38:1124 – 44.
38. Stacpoole PW, Wright EC, Baumgartner TG, et al. A controlled clinical trial of dichloroacetate for treatment of lactic acidosis in adults. N Engl J Med. 1992;327:1564 –9.
39. Alberti KGMM, Holloway PAH. Dichloroacetate and phenformin- induced lactic acidosis. Diabetes 1977;26:377.
40. Stacpoole PW, Lorenz AC, Thomas RG, Harman EM. Dichloro- acetate in the treatment of lactic acidosis. Ann Intern Med. 1988; 108:58 – 63.